CDC conducts weekly surveillance for laboratory-confirmed COVID-19 and influenza-associated hospitalization rates with a catchment area covering 10% of the U.S. population. However, case reporting delays, particularly for the most recent weeks, create difficulty in communicating rates. We developed a method to accurately predict recent hospitalization rates, while accounting for expected reporting delay patterns. Generally, for both viruses, 90% of hospitalized cases are ascertained within two weeks of hospital admission, but this delay varies by surveillance site and over time. We took a multiplier approach to adjust COVID-19 hospitalization rates based on historical trends of reporting delays by site from March 2020 to March 2021 and compared with an established NowCast Bayesian Smoothing model. We tested both methods on hospitalization rates at various time points. Both approaches usually predicted the “true” hospitalization rate (i.e., rate observed after >99% of cases captured), within +/-5%. We next plan to test both methods on historical influenza hospitalization data, describe periods when these approaches deviate most from the “truth”, and describe the pros and cons.